INDEX

Explanations

number strings

oai_token-act-pair · gpt-3.5-turbo

references to numerical data and names associated with entities or terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 departures

-0.62

 millenn

-0.57

 conclud

-0.56

 Lauder

-0.56

 tremend

-0.55

Ö¼

-0.55

FML

-0.53

 unification

-0.50

 nomine

-0.50

 protracted

-0.50

POSITIVE LOGITS

thur

0.68

sels

0.64

utra

0.63

ucket

0.58

 brim

0.56

iasco

0.56

 Splash

0.53

agall

0.53

 Refresh

0.52

tis

0.52

Activations Density 1.309%

number strings

references to numerical data and names associated with entities or terms

No Comments

No Known Activations